Add Your Heading Text Here
Add Your Heading Text Here
Add Your Heading Text Here
Add Your Heading Text Here
Add Your Heading Text Here
Add Your Heading Text Here
Add Your Heading Text Here

Sanat Thukral

Project Title

Explainable auto-regressive autoencoder models of real-time mental workload assessment for E-learning with EEG data using passive brain-computer interfaces.

Project Description

AI and ML advancements have led to models for learning complex patterns in EEG data and brain activation [3,7]. These models contribute to understanding human behavior and brain responses, motivating the design of passive brain-computer interfaces (pBCIs) for creating novel adaptive and personalized technologies. pBCIs enhance human-computer interactions by facilitating communication channels between the brain and external devices [5]. A critical use case is estimating mental workload (MWL) to develop user-centered applications, with EEG data often used [1],[2]. Unsupervised autoencoders can learn complex patterns in EEG data, but they are black-boxes, making interpretation and explanation difficult. This research aims to create explainable models for MWL estimation using EEG data in E-learning [3]. The model will offer insights into cognitive processes and mental workload variations experienced by students on an E-learning platform [3]. Variations in taskload will be induced by manipulating the perceptual modality from Multiple Resource Theory [8] in the E-learning interface, expecting different mental workload levels. In summary, this study contributes to the body of knowledge by developing a real-time MWL model induced from EEG data for E-learning contexts. This model will underpin the construction of pBCIs that facilitate the design of human-centered technologies aligned with brain information processing.

Works Cited: 1. Longo, L. (2022). Modeling cognitive load as a self-supervised brain rate with electroencephalography and deep learning. *Brain Sciences*, *12*(10), 1416.

2. Raufi, B., & Longo, L. (2022). An Evaluation of the EEG alpha-to-theta and theta-to-alpha band Ratios as Indexes of Mental Workload. *Frontiers in Neuroinformatics*, *16*, 44.

3. Das, A., & Rad, P. (2020). Opportunities and challenges in explainable artificial intelligence (xai): A survey. *arXiv preprint arXiv:2006.11371*. https://arxiv.org/pdf/2006.11371.pdf

4. Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., Chatila, R., & Herrera, F. (2020). Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58, 82–115. https://doi.org/10.1016/j.inffus.2019.12.012

5. Hancock, G. m., Longo, L., Young, M. s., & Hancock, P. a. (2021). Mental Workload. In HANDBOOK OF HUMAN FACTORS AND ERGONOMICS (pp. 203–226). John Wiley & Sons, Ltd. https://doi.org/10.1002/9781119636113.ch7

6. Longo, L. (2014). Formalising Human Mental Workload as a Defeasible Computational Concept [Thesis, Trinity College Dublin]. http://www.tara.tcd.ie/handle/2262/72197

7. Moustafa, K., Luz, S., & Longo, L. (2017). Assessment of Mental Workload: A Comparison of Machine Learning Methods and Subjective Assessment Techniques. In L. Longo & M. C. Leva (Eds.), Human Mental Workload: Models and Applications (pp. 30–50). Springer International Publishing. https://doi.org/10.1007/978-3-319-61061-0_3

8. Wickens, C. D. (2008). Multiple Resources and Mental Workload. *Human Factors*, *50*(3), 449–455. https://doi.org/10.1518/001872008X288394